This paper proposes a practical method for identifying wheeling paths in deregulated electricity markets based on an extended sensitivity analysis. Using this method, it becomes possible to decide the proper and fair wheeling rate according to the degree of burden on transmission lines by each power flow transaction. Moreover, a wheeling rate based on the real power flow burden is also an important signal to new power suppliers in the markets. In order to show the validity of the proposed method, a series of simulations on the IEEE 30-bus test system were conducted.
There is a trend toward competitive markets in the electric power industry all over the world [1], [2] and [3]. In Japan, although the deregulation of electricity is being implemented mainly in the wholesale power market, partial retail wheeling is allowed. In other words, any party (player) who is in the business of supplying electricity to the market can supply it to its customers, not only from its own generators but also from the market, selling both inside and outside their regions. High voltage customers can purchase power from either existing regional utility companies, or new power suppliers. Since the pricing for these purchases is not regulated, bilateral contracts between suppliers and customers may be negotiated and arranged. However, if electricity based on the bilateral transactions uses the grid owned by the regional utility company, the wheeling rate must be paid to the grid owner. Under such circumstances, appropriate setting of the wheeling rate is one of the critical issues for both suppliers and grid owners in deregulated electricity markets [4] and [5].
Several methodologies to determine wheeling rates, such as postage stamp, contract path, distance based MW-mile, and power flow based MW-mile have been proposed and deeply investigated [6], [7] and [8]. However, since in the aforementioned methods, the use of transmission lines and time variable load flows is not modeled sufficiently, a wheeling rate reflecting the transmission line condition (change by time and a day of week) is preferable in order to obtain a fair rate.
In this paper, we propose an efficient method for identifying wheeling paths based on an extended sensitivity analysis. Using this method, it becomes possible to fix the proper and fair wheeling rate according to the degree of burden on transmission lines by each power flow transaction. The proposed method has four advantages compared with the existing transmission routes identification techniques. (i) The proposed approach incorporates the concept of generation distribution factor [9], [10] and [11] into the sensitivity analysis, which can take the consideration of not only variable loads but also generator characteristics (generation capacity, speed regulation, and dispatching strategies) and load characteristics (voltage and frequency elasticity, constant power features) under various conditions. (ii) Comparing with most of the techniques which target to identify wheeling path from single generator to single customer [12], [13] and [14], the proposed method can identify multiple wheeling paths accurately, i.e., any combination among the generators and the customers, such as single to single, plurality to single, single to plurality and plurality to plurality. This approach is applicable to the situation where new power suppliers have more than two generators and there are many customers in the market. (iii) The computational complexity of the proposed method is approximately equivalent to one iteration of power flow calculation, and is computationally efficient. (iv) There is no special assumption, and also no additional theoretical error except numerical error.
The paper is organized as follows. In the next section, the sensitivity and generation distribution factor is described. In Section 3, in order to identify the wheeling paths of any combination among generators and customers, the sensitivity factor concept introduced in Section 2 is expanded. Then, wheeling paths identification technique and the algorithm that uses the extended sensitivity method are described. Moreover, we explain the outline of applications of the proposed method to the actual power market. In Section 4, in order to show the validity of the proposed method, several simulations on the IEEE 30-bus test system are conducted and numerical results are described. Section 5 concludes the paper.
